Remove calls to styleForLayoutObject() in LayoutTreeBuilder::style()

Remove calls to styleForLayoutObject() in LayoutTreeBuilder::style()

This patch stashes all StyleReattachData information about all
descendants when an element will need to undergo Reattach.
It does so by recursing over all ContainerNode, ShadowRoot
and PseudoElement descendants in the case where the node 
itself needs to Reattach.

We then use the ComputedStyle stashed on the HashMap in
LayoutTreeBuilder::style() eliminating the need to call
Element::styleForLayoutObject() in Layout Tree Construction.

BUG=695267, 595137

[nainar, 2017-03-03 06:08:51]

nainar@chromium.org changed reviewers:
+ esprehn@chromium.org

[nainar, 2017-03-03 06:08:52]

Elliott, 

No action required on this. More of an FYI on the progress.

https://codereview.chromium.org/2727233005/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/dom/Element.cpp (right):

https://codereview.chromium.org/2727233005/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/dom/Element.cpp:2030:
recalcShadowRootStylesForLayout();
This only goes over the descendants (ContainerNodes and ShadowRoot descendatnts)
only in the case we need Reattach. Should we maybe doing this in cases where
change >= UpdatePseudoElements to mimic the code in Element::recalcStyle()?

https://codereview.chromium.org/2727233005/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/dom/Element.cpp:2082:
recalcShadowRootStylesForLayout();
Need to add a function here to deal with the relevant PseudoElement as well. 

void Element::recalcPseudoStyleForLayout(PseudoId pseudoId) {
  if (PseudoElement* element = pseudoElement(pseudoId))
      element->recalcStyleForLayout();
}

This comment applies to the code change above as well.

[nainar, 2017-03-03 06:21:41]

The CQ bit was checked by nainar@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-03 06:21:56]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-03 07:02:51]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-03 07:02:52]

Dry run: Try jobs failed on following builders:
  linux_chromium_tsan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[nainar, 2017-03-09 02:44:28]

The CQ bit was checked by nainar@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-09 02:44:46]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[nainar, 2017-03-09 02:50:19]

Description was changed from

==========
Remove calls to styleForLayoutObject() in LayoutTreeBuilder::style()

This patch stashes all StyleReattachData information about all
descendants when an element will need to undergo Reattach.
It does so by recursing over all ContainerNode and ShadowRoot
descendants in the case where the node itself needs to Reattach.

We then use the ComputedStyle stashed on the HashMap in
LayoutTreeBuilder::style() elimitating the need to call
Element::styleForLayoutObject() in Layout Tree Construction.
BUG=695267
==========

to

==========
Remove calls to styleForLayoutObject() in LayoutTreeBuilder::style()

This patch stashes all StyleReattachData information about all
descendants when an element will need to undergo Reattach.
It does so by recursing over all ContainerNode, ShadowRoot
and PseudoElement descendants in the case where the node 
itself needs to Reattach.

We then use the ComputedStyle stashed on the HashMap in
LayoutTreeBuilder::style() eliminating the need to call
Element::styleForLayoutObject() in Layout Tree Construction.

BUG=695267, 595137
==========

[nainar, 2017-03-09 02:53:04]

Just uploading code from yesterday. 

Four failing fast tests:
fast/css-generated-content/016.html

Minimized test case:
<style>
	a:after {
		content: '(href=' attr(href) ')';
	}
</style>
<p><a href="1">1</a></p>
<p><a href="2">2</a></p>


fast/css-generated-content/attr-minus.html

No minimization needed

fast/css-generated-content/pseudo-animation-before-onload.html

<style>
@keyframes example {
  from { width: 50px; height: 50px; }
  to { width: 10px; height: 10px; }
}

.before:before {
    content: "";
    display: block;
    position: relative;
    background: green;
    width: 10px;
    height: 10px;
    animation: example 2s;
}

.before {
    display: inline-block;
}

</style>

<div class="before"></div>
<div class="before"></div>

fast/css/invalidation/in-and-out-of-range-pseudo.html

Working on minimizing. Please note I would suspect the first three test failures
are related.

https://codereview.chromium.org/2727233005/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/dom/Element.cpp (right):

https://codereview.chromium.org/2727233005/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/Element.cpp:29: #include <memory>
git cl format. Will get rid of this.

https://codereview.chromium.org/2727233005/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/Element.cpp:114: #include
"core/html/HTMLObjectElement.h"
Not needed.

[commit-bot: I haz the power, 2017-03-09 03:51:38]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-09 03:51:39]

Dry run: Try jobs failed on following builders:
  linux_chromium_chromeos_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[nainar, 2017-03-09 06:54:12]

The CQ bit was checked by nainar@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-09 06:54:32]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-09 07:52:17]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-09 07:52:18]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[alancutter (OOO until 2018), 2017-06-30 06:37:16]

alancutter@chromium.org changed reviewers:
+ alancutter@chromium.org

[alancutter (OOO until 2018), 2017-06-30 06:37:17]

If this patch lands we can land a fix to a very old transition bug:
https://chromium-review.googlesource.com/c/557586/